A method for detecting deviations between an event log and a process model includes converting the process model into a probability process model, the probability process model comprising multiple nodes in multiple hierarchies and probability distribution associated with the multiple nodes, a leaf node among the multiple nodes corresponding to an activity in the process model; detecting differences between at least one event sequence contained in the event log and the probability process model according to a correspondence relationship; and identifying the differences as the deviations in response to the differences exceeding a predefined threshold; wherein the correspondence relationship describes a correspondence relationship between an event in one event sequence of the at least one event sequence and a leaf node in the probability process model.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
4. The method according to claim 3, wherein the path comprises a multi-level path corresponding to multiple hierarchies in the probability process model.
This invention relates to probabilistic modeling and data processing, specifically improving the efficiency and accuracy of path-based computations in hierarchical probability process models. The problem addressed is the computational complexity and scalability challenges when analyzing multi-level hierarchical structures in probabilistic models, such as Bayesian networks or Markov chains, where dependencies span multiple layers. The method involves constructing and analyzing a multi-level path within a probability process model, where the path corresponds to multiple hierarchies or nested structures in the model. This allows for more efficient traversal and evaluation of probabilistic dependencies across different levels of the hierarchy, reducing redundant computations and improving accuracy. The path may include intermediate nodes or states that represent transitions or dependencies between higher and lower levels of the hierarchy, enabling more precise modeling of complex relationships. The method can be applied to various probabilistic models, including those used in machine learning, statistical analysis, and decision-making systems. By leveraging the multi-level path structure, the approach optimizes the evaluation of probabilities, likelihoods, or other statistical measures, particularly in scenarios where hierarchical dependencies are critical. This enhances computational efficiency and scalability, making it suitable for large-scale or high-dimensional probabilistic models.
12. The computer program product of claim 11, wherein the path comprises a multi-level path corresponding to multiple hierarchies in the probability process model.
A system and method for analyzing probability process models involves generating a path representing a sequence of states or events within the model. The path is constructed by sampling from the model, where each state transition is determined based on transition probabilities defined by the model. The system further includes a visualization component that displays the path, allowing users to observe the progression of states over time. The path may include multiple levels, corresponding to different hierarchies within the model, where each level represents a distinct layer of abstraction or granularity in the process. For example, a higher-level path may represent broad phases of the process, while lower-level paths may detail individual steps within those phases. The visualization may highlight key transitions, probabilities, or other relevant metrics to aid in understanding the model's behavior. This approach enables users to analyze complex probabilistic processes by breaking them down into interpretable paths, facilitating decision-making, debugging, or optimization of the underlying model. The system may be implemented as a software tool, integrating with existing modeling frameworks or databases to extract and process the necessary data.
20. The system according to claim 19, wherein the path comprises a multi-level path corresponding to multiple hierarchies in the probability process model.
The system involves a probabilistic modeling framework designed to analyze and predict outcomes based on hierarchical data structures. The core challenge addressed is the efficient representation and processing of complex, multi-level dependencies in data, particularly where relationships span multiple hierarchical layers. Traditional models often struggle with such structures, leading to inaccuracies or computational inefficiencies. The system includes a probability process model that captures dependencies across different levels of hierarchy, enabling more accurate predictions. A key feature is the use of a multi-level path within this model, which allows the system to traverse and analyze relationships across multiple hierarchical layers simultaneously. This path-based approach ensures that dependencies between different levels are properly accounted for, improving the model's ability to handle nested or layered data structures. The system may also include components for generating, storing, and updating the probability process model, as well as mechanisms for processing input data to extract relevant features and relationships. By integrating these elements, the system provides a robust framework for modeling and predicting outcomes in domains where hierarchical dependencies are critical, such as organizational structures, biological systems, or networked data. The multi-level path feature enhances flexibility and accuracy, making the system suitable for applications requiring deep hierarchical analysis.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 6, 2019
November 29, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.